In what follows I would like to examine the objections to Molinism and its concept of middle knowledge by Paul Helm a prominent Calvinist philosopher of religion. Even though I have many points of disagreement with Helm, I really enjoy reading his work. In dealing with Helm’s objections we will survey and explore the foundational differences in Helm’s view in light of the explanation of Molinism provided here. We will then be able to responsibly critique Helm’s reasons for rejecting what Dr. William Lane Craig and others regard as “one of the most fruitful theological ideas ever conceived,”1 namely middle knowledge, and we will split the post up into two articles. In this first article we will deal with Helm’s first objection to middle knowledge and then deal with the second objection he raises in the next article.
Possible Worlds and Feasible Worlds
Important for our discussion is the difference in possible worlds and feasible worlds.2 In God’s natural knowledge God knows all possible worlds, such as “Chan would freely buy a Dodge Challenger Hellcat if he had the money,” and “Chan would freely not buy a Dodge Challenger Hellcat if he had the money.” Both of those propositions would be part of God’s natural knowledge. In the next logical moment, God knows the truth value, given human freedom, of what Chan would in fact do. So if it is true that “Chan would freely buy a Dodge Challenger Hellcat if he had the money,” then the world in which he would not is not feasible for God to bring about given human freedom. So God cannot bring it about that Chan will freely not buy a Dodge Challenger Hellcat if he had the money because it requires denying Chan indeterministic freedom. God, via His middle knowledge can choose a world to create and guide it providentially to achieve His ultimate ends despite indeterministic human freedom.
Considering our explication of Molinism in a previous article, it would be what Paul Helm would call a ‘no-risk’ view of God’s providence. In his work The Providence of God Helm states the following leading into discussing middle knowledge,
“So far, we have presented the ‘risk’ view of providence (resting on an indeterministic view of human freedom) and the ‘no-risk’ as exclusive alternatives. But they may not be. It may be possible to combine the strong view of human freedom and a ‘no-risk’ view of divine providence. If such a combination is possible, it would in the eyes of many represent the best of all possible outcomes; a strong view of providence and a strong, indeterministic view of human freedom.” 3
Helm’s Objections to Middle Knowledge
After offering that there may be a solution to the seeming exclusivistic notions of indeterministic human freedom and a ‘no-risk’ view of providence Helm launches into a description of Molinism and then a critique concluding that as seductive as it may be, this view ultimately cannot provide what it promises.4 Helm offers two major objections:
1. Indeterministic freedom cannot be reconciled with God’s foreknowledge.5
2. The Molinist appeal to mystery in regard to the truth of counterfactuals is no different than the Augustinian-Calvinist appeal to mystery in regard to God as the author of evil.6
Objection 1: Divine Foreknowledge and Indeterministic Freedom are Incompatible.
This objection by Helm is the classic objection that if a choice is truly indeterministic then by the nature of what it is it cannot be known until the time the choice is made. This would make it impossible for God to know what choice a person would make since they have the option of several choices being inderministically free. Since God’s knowledge is infallible then if God knows what choice the person will make then that person is not free to do otherwise and will therefore choose what God knows they will choose.
Understood this way, if an agent is indeterministically free then that person is free to choose whatever they want right up until they do in fact make a choice. What is the Molinist to make of this objection? This actually ends up being a common error in modal logic made by both open theists as well as Augustinian-Calvinists. As William Lane Craig commenting on Helm’s essay states, “What strange bedfellows these openness and Reformed theologians be!”7 The error is to confuse necessity with certainty in relation to God’s knowledge of future contingents. This objection is commonly raised by arguing as follows:
1. Necessarily, if God knows x, then x will happen
2. God foreknows x
3. Therefore x will happen necessarily8
This common form of the argument is logically fallacious and Helm knows this. To have a necessary conclusion modal logic requires that both premises be necessary as well. Helm argues as follows so his argument is exempt from the logical blunder of the common form of the argument for theological fatalism using the example of Jones eating a tuna sandwich:9
1. Necessarily, if God foreknows x, then x will happen.
2. Necessarily, God foreknows x.
3. Therefore, x will necessarily happen10
Using the argument above, but replacing x with “Jones eating a tuna sandwich,” we can see the argument is now logically valid. The Molinist would of course challenge the truth of premise 2, for why think God’s foreknowledge of an event x is necessary? This would mean that God had to create the actual world necessarily! In other words, God had no options when it comes to creation of the actual world! As William Lane Craig observes,
“Christian theologians have always insisted that the content of God’s foreknowledge is not necessary. He could have created a world different than this one or no world at all, in which case his foreknowledge would be different. Necessarily, whatever God knows is true, but is not necessary that God know what he knows. To say God’s foreknowledge is necessary is to say that this is the only world he could have created and that he creates it necessarily.”11
Helm tries to avoid the charge of fatalism by introducing what he calls “the necessity of the past.” Helm states,
“If there is something in the past that entails something in the future and if what is past is necessary – accidentally or historically necessary – then what is entailed is similarly accidentally or historically necessary.”12
Craig deals with this argument at length13 working through the issues such as changing versus causing the past, the difference that one’s theory of time makes,14 etc. He concludes with this observation,
“no fatalist has, to my knowledge, explained the necessity of the past as anything other than either the unalterability or the causal closedness of the past,”15
This entails that Helm’s adding the “necessity of the past” objection then fails in getting him out of the fatalist woods so to speak. Next time we will explore Helm’s second objection to middle knowledge, namely that he claims the Molinist appeal to mystery in regard to the truth of counterfactuals is no Different than the Augustinian-Calvinist appeal to mystery in regard to God as the author of evil.